Indian Palaeography from about B.C. 350 to about A.D. 1300 c7.s9.13

c7.s9.13

A second peculiarity,¹ found in many inscriptions in Prākrit and in the mixed dialect, is the frequent erroneous employment of the signs for the sibilants. In the Aśoka edicts of Kālsī, of Siddāpura, and of Bairāt No. II,² on the Bhaṭṭiprolu vases, in the cave inscriptions of Nāgārjunī and of Rāmnāth,³ and in the Mathurā inscriptions of the Kuṣana period, nay even in the two oldest Ceylonese inscriptions, *ṣa* or *śa* are used often for *sa*, and *śa* for *ṣa*, and *sa* for *śa* and *ṣa*. The reasons for this promiscuous use of the sibilants are, first, the circumstance that the school alphabet, which the clerks learned, was originally intended for Sanskrit and contained more sibilants than the ancient vernaculars possessed, and secondly, the negligent pronunciation of the classes destitute of grammatical training. The western and southern Prākrits very probably possessed, then as now, both the palatal and the dental sibilants, and it was probably the custom, as is done also in our days, to exchange the two sounds in the same words. The natural consequence was that the feeling for the real value of the signs for *śa* and *sa* disappeared among the Prākrit-speaking classes, while the *ṣa* of their school-alphabet, for which there was no corresponding sound in their vernaculars, must have appealed to them as a sign suitable to express sibilance. The Sanskrit inscriptions of all centuries, especially the land-grants which were drafted by common clerks, the MSS. of works written in the modern Prākrits, and the documents from [32] the offices of modern India, with their countless mistakes in the use of the sibilants, offer abundant proof for the correctness of this explanation of the errors in the old inscriptions. The explanation is also confirmed by the occasional occurrence of *ṇa*⁴ for *na*, — once in the separate edicts of Dhauli and once of Jaugada, — though *na* alone is permissible for their dialect. In these cases, too, the error seems to have been caused by the fact that the school-alphabet contained both *ṇa* and *na*. The clerks, who had learned it, each made once a slip, and put in the, for them, redundant sign. The different opinion,⁵ according to which the exchange of the sibilants in the Aśoka edicts indicates that the values of the Brāhma signs were not completely settled in the third century B. C., rests on the, now untenable, assumption that the Brāhmī was elaborated, not for writing Sanskrit, but for the Prākrit dialects.